The work develops an innovative method to recognize bipolar disorder patient depression patterns through the combined use of Patient Health Questionnaire-9 (PHQ-9) and Montgomery Asberg Depression Rating Scale (MADRS) decision tools. The proposed system combines wearable sensors with machine learning capabilities to analyze depressive symptoms which involve ongoing sadness together with loss of pleasure and sleep issues and appetite and energy level changes and mental dysfunction. The acquisition of data happens through smartwatches that record constant measurements of heart rate variability (HRV) along with resting heart rate and activity levels. The gathered information goes through K-Means clustering analysis to provide depression severity classifications among individuals. The framework utilizes five clusters to classify results from the PHQ-9 into depression levels starting from Minimal depression (0–4) which may not need treatment to severe depression (20–27) requiring intensive therapeutic intervention. The proposed assessment system provides both whole-time and non-invasive check-up capabilities which help practitioners intervene early followed by individualized therapeutic methods. This research gains worldwide applicability because the PHQ-9 assessment tool provides free access in more than thirty languages representing diverse populations.